Generate images and video right inside Claude Code, Codex, Cursor or Cherry Studio

Fastest: the Codex / Claude Code plugin

Coding agents such as Codex and Claude Code are good at planning: scripts, shot lists, prompts, editing with ffmpeg. The apimodels plugin adds the step they cannot do — actually rendering the video, images and voice-over. It bundles this MCP server with a skill that tells the agent when to use it, how to pick a model, how long video takes and when to ask you before spending. Once installed, just ask in plain words; there is no need to name tools.

1. 1. Sign in once

   Your browser opens apimodels.app; click Authorize. A dedicated key is created for this machine and stored locally — nothing to copy or paste — and every client (Codex, Claude Code, …) picks it up. The agent can run this step for you; you only click Authorize.

   Copy

   npx -y apimodels-mcp login

2. 2. Install the plugin

   Claude Code(2.1.275+):

   Copy

   /plugin install apimodels --marketplace stimQQ/apimodels-mcp

   Codex (restart Codex afterwards):

   Copy

   codex plugin marketplace add stimQQ/apimodels-mcp
   codex plugin add apimodels@apimodels

   Only have the Codex desktop app and no codex command? Replace codex with npx -y @openai/codex — it runs the official CLI on the fly and shares the desktop app's config.

3. 3. Just ask

   For example "make a 5-second 16:9 video of a kitten slowly raising its head in morning light" or "plan a 20-second product video for this repo and render it shot by shot". Codex asks you to confirm calls that spend credits (video, image, speech); checking progress or the balance never asks.

The plugin and the MCP server need Node.js. If it is missing, install the plugin anyway and ask for something: the agent notices the tools are not loaded and, with your OK, installs the official Node.js for you (winget on Windows — click Yes once; Apple's installer on macOS — click Continue, then Install). Restart the agent afterwards.

Hosted server: no Node.js, one Authorize click

The same tools also run on our servers. Add this URL to your client; it asks you to sign in to apimodels.app and click Authorize, and you are done — no Node.js, no key to copy. Authorizing creates a dedicated key named “MCP · app · date”; delete it in the console to disconnect.

https://api.apimodels.app/mcp

# ~/.codex/config.toml
[mcp_servers.apimodels]
url = "https://api.apimodels.app/mcp"

# then, in a terminal:
codex mcp login apimodels

claude mcp add --transport http apimodels https://api.apimodels.app/mcp
# then in Claude Code: /mcp → apimodels → Authenticate

{
  "mcpServers": {
    "apimodels": { "url": "https://api.apimodels.app/mcp" }
  }
}

Two differences from the local server: the hosted one cannot read files on your machine — the agent calls get_upload_url, uploads the file with curl and uses the URL it gets back (a local path is refused with that hint); and each call waits at most 50 seconds, handing back a task id for get_task if a video is still rendering. If you already have a key, you can also connect with an Authorization: Bearer sk_… header.

Install

You can also configure the server yourself. There is nothing to download — it runs on demand through npx. Either run the login command above once (then the env line in the configs below can go), or grab a key at Console → API Keys and put it in.

{
  "mcpServers": {
    "apimodels": {
      "command": "npx",
      "args": ["-y", "apimodels-mcp"],
      "env": {
        "APIMODELS_API_KEY": "sk_your_key_here"
      }
    }
  }
}

{
  "mcpServers": {
    "apimodels": {
      "command": "npx",
      "args": ["-y", "apimodels-mcp"],
      "env": { "APIMODELS_API_KEY": "sk_your_key_here" }
    }
  }
}

[mcp_servers.apimodels]
command = "npx"
args = ["-y", "apimodels-mcp"]
env = { APIMODELS_API_KEY = "sk_your_key_here" }
# tool_timeout_sec = 660   # optional, see "Long generations" below

Any other MCP client works the same way — run this over stdio.

APIMODELS_API_KEY=sk_your_key_here npx -y apimodels-mcp

npx needs Node.js on your machine. On Windows install it from nodejs.org — Cherry Studio relies on it to launch the MCP server too.

What the server exposes

ElevenLabs TTS is not exposed here: it streams raw audio bytes from POST /v1/tts/stream rather than returning a URL. Use the audio API directly for it.

Let the model check its own work

You can say "make a 16:9 banner that says SAVE 10%, check the spelling and fix it if it is wrong" and let the assistant iterate until it is right. It sees the result through one of two paths, and which one applies depends on your client:

Reviews default to gpt-5.6-luna and cost well under $0.01 each; pass claude-sonnet-5 for a stricter read. The conversation model itself only needs tool-calling support — Claude, GPT-5.x, gpt-6-astra and Gemini all qualify. You never fill in aspect ratio or resolution: the assistant picks them from what you asked for, so saying "16:9" is enough.

Local images just work

image_url on generate_image and generate_video accepts four forms: a public https:// URL, a local file path (/Users/me/photo.png, ./ref.jpg, ~/Pictures/x.webp), a URL on your own machine (http://127.0.0.1:8000/photo.png), and a data:image/png;base64,… URI.

The last three are uploaded for you first, and the resulting public URL is what gets generated from. That step has to happen on your side: the file exists only on your machine, and 127.0.0.1 means our server when our server resolves it — which is why passing one straight to the REST API fails with "private/reserved IP addresses not allowed". Uploads land in your account storage and are deleted after 30 days.

Long generations: nothing gets lost

Every generation on apimodels is an async task, and video is slow: a median of about 2.5 minutes, 9 in 10 within 8 minutes, on our production traffic for the week to 2026-09-22 (images take about 50 seconds). Meanwhile Codex aborts a tool call after 60 seconds by default, and the MCP SDK's own client timeout is 60 seconds too. A tool that blocks until the video is ready gets killed mid-wait: the task completes and is billed, but the assistant never sees the URL. That is what 0.2.x did.

Since 0.3.0 the generation tools wait at most wait_seconds (default 50) and then return the task id with a "still running" note; the assistant calls get_task, which waits up to another wait_seconds and returns the URL(s) — with the preview for image tasks — or "still running" again. Nothing is resubmitted and nothing is billed twice, and the assistant handles it by itself. Claude Code, Claude Desktop and Cursor have no 60-second limit, so wait_seconds: 600 on generate_video gets the URL in one call; APIMODELS_WAIT_SECONDS=600 in the server's env makes that the default.

Configuration

Common pitfalls